Deep Unsupervised Learning for Joint Antenna Selection and Hybrid Beamforming
- 14 January 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 70 (3) , 1697-1710
Abstract
In this paper, we propose a novel deep unsupervised learning-based approach that jointly optimizes antenna selection and hybrid beamforming to improve the hardware and spectral efficiencies of massive multiple-input-multiple-output (MIMO) downlink systems. By employing ResNet to extract features from the channel matrices, two neural networks, i.e., the antenna selection network (ASNet) and the hybrid beamforming network (BFNet), are respectively proposed for dynamic antenna selection and hybrid beamformer design. Furthermore, a deep probabilistic subsampling trick and a specially designed quantization function are respectively developed for ASNet and BFNet to preserve the differentiability while embedding discrete constraints into the network structures. With the aid of a flexibly designed loss function, ASNet and BFNet are jointly trained in a phased unsupervised way, which avoids the prohibitive computational cost of acquiring training labels in supervised learning. Simulation results demonstrate the advantage of the proposed approach over conventional optimization-based algorithms in terms of both the achieved rate and the computational complexity.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (61831013)
- Tsinghua University Initiative Scientific Research Program
- Tsinghua University-China Mobile Communications Group Co., Ltd. Joint Institute
- Science and Technology Commission Foundation of Shanghai (20DZ1101200)
This publication has 48 references indexed in Scilit:
- Alternating Minimization Algorithms for Hybrid Precoding in Millimeter Wave MIMO SystemsIEEE Journal of Selected Topics in Signal Processing, 2016
- Hybrid Digital and Analog Beamforming Design for Large-Scale Antenna ArraysIEEE Journal of Selected Topics in Signal Processing, 2016
- Hybrid MIMO Architectures for Millimeter Wave Communications: Phase Shifters or Switches?IEEE Access, 2016
- Convolutional neural networks at constrained time cost2015
- Hybrid beamforming with finite-resolution phase shifters for large-scale MIMO systems2015
- Large-scale antenna systems with hybrid analog and digital beamforming for millimeter wave 5GIEEE Communications Magazine, 2015
- MIMO Precoding and Combining Solutions for Millimeter-Wave SystemsIEEE Communications Magazine, 2014
- Efficient Transmit Antenna Selection for Multiuser MIMO Systems with Block Diagonalization2007
- Fast Antenna Subset Selection in MIMO SystemsIEEE Transactions on Signal Processing, 2004
- Feature selection, L1 vs. L2 regularization, and rotational invariance2004